196 C. E. Beecher—The Morphology of Triarthrus. 



animal may be recognized by the earlier or later appearance of their 

 rudiments." 1 



In Triarthrus, these disturbing factors are hardly to be recognized, 

 for no pair of limbs had an excessive physiological importance over 

 any other pair or series of pairs, and increase progressed regularly 

 by the addition of new members in front of the anal segment. The 

 pygidium, being formed of fused segments, accommodated itself to 

 this kind of growth by pushing forward the series of limbs and by 

 the formation of a new free segment at the posterior end of the 

 thorax. This process of metameric growth continued from the 

 protaspis stage with no free thoracic segments, and successively 

 added segment after segment with corresponding moults, until the 

 full complement was reached, after which the moulting resulted 

 mainly in increase in size. The repetition of moults afforded the 

 chief means by which modifications in the appendages could be 

 brought about. 



The earliest protaspis stage shows, from the segmentation of the 

 axis, that there were present five pairs of appendages on the head 

 and two on the pygidium (6). The adult animal has thirteen or 

 fourteen free thoracic segments and six pygidial. 2 Now, so far as is 

 known of trilobite ontogeny, there was nevermore than one segment 

 added at a single moult, though there is no evidence that there may 

 not have been more moults than segments between the protaspis 

 stage and the finished segmentation. In Triarthrus, the average full 

 number of segments was attained by the time the animal reached 

 a length of about 7 mm. : so that the limbs of the anterior thoracic 

 segment in an individual 7 mm. in length, and containing the full 

 complement of fourteen free and six pygidial segments, must have 

 undergone at least seventeen moults. The second thoracic segment, 

 therefore, at this stage of growth would have been moulted sixteen 

 times, the fifth thirteen times, the tenth eight times, and the four- 

 teenth four times. The length of full-grown individuals is from 25 

 to 40 mm., and to have reached this size a considerable number of 

 additional moults must have occurred, in which all the segments 

 participated alike. 



Some mention should be made of the probable method of 

 respiration of Triarthrus. No traces of any special organs for this 

 purpose have been found in this genus, and their former existence 

 is very doubtful, especially in view of the perfection of details 

 preserved in various parts of the animal. 



The delicacy of the appendages and ventral membrane of 

 trilobites and their rarity of preservation are sufficient demonstration 

 that these portions of the outer integument were of extreme 

 thinness, and therefore perfectly capable of performing the function 

 of respiration. Similar conditions occur in most of the Ostracoda 

 and Copepoda, and also in many of the Cladocera and Cirrepedia, 

 where no special respiratory organs are developed. 



1 "Textbook of Comparative Anatomy," English edition (Bernard), p. 410. 



2 A few individuals of this species (T. Becki) have been observed with one or two 

 additional thoracic segments. — Walcott (11). 



